Twitter sentiment analysis

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 74682 entries, 0 to 74681
Data columns (total 4 columns):
 #   Column  Non-Null Count  Dtype 
---  ------  --------------  ----- 
 0   ID      74682 non-null  int64 
 1   Entity  74682 non-null  object
 2   Label   74682 non-null  object
 3   Text    73996 non-null  object
dtypes: int64(1), object(3)
memory usage: 2.3+ MB

1: im getting on borderlands and i will murder you all , -> Positive
2: I am coming to the borders and I will kill you all, -> Positive
3: im getting on borderlands and i will kill you all, -> Positive
4: im coming on borderlands and i will murder you all, -> Positive
5: im getting on borderlands 2 and i will murder you me all, -> Positive
6: im getting into borderlands and i can murder you all, -> Positive
7: So I spent a few hours making something for fun. . . If you don't know I am a HUGE @Borderlands fan and Maya is one of my favorite characters. So I decided to make myself a wallpaper for my PC. . Here is the original image versus the creation I made :) Enjoy! pic.twitter.com/mLsI5wf9Jg -> Positive
8: So I spent a couple of hours doing something for fun... If you don't know that I'm a huge @ Borderlands fan and Maya is one of my favorite characters, I decided to make a wallpaper for my PC.. Here's the original picture compared to the creation I made:) Have fun! pic.twitter.com / mLsI5wf9Jg -> Positive
9: So I spent a few hours doing something for fun... If you don't know I'm a HUGE @ Borderlands fan and Maya is one of my favorite characters. -> Positive
10: So I spent a few hours making something for fun. . . If you don't know I am a HUGE RhandlerR fan and Maya is one of my favorite characters. So I decided to make myself a wallpaper for my PC. . Here is the original image versus the creation I made :) Enjoy! pic.twitter.com/mLsI5wf9Jg -> Positive

Shape of X_train:  (59196,)
Shape of X_test:  (14800,)

0.7322972972972973

              precision    recall  f1-score   support

           0       0.95      0.46      0.62      2575
           1       0.65      0.90      0.76      4472
           2       0.84      0.64      0.72      3622
           3       0.71      0.81      0.76      4131

    accuracy                           0.73     14800
   macro avg       0.79      0.70      0.71     14800
weighted avg       0.77      0.73      0.72     14800

0.9118918918918919

              precision    recall  f1-score   support

           0       0.97      0.86      0.91      2575
           1       0.92      0.93      0.93      4472
           2       0.94      0.89      0.91      3622
           3       0.85      0.94      0.90      4131

    accuracy                           0.91     14800
   macro avg       0.92      0.91      0.91     14800
weighted avg       0.91      0.91      0.91     14800